A compact record of one tool invocation with its input and output behind a disclosure.
A compact record of one tool invocation with its input and output behind a disclosure.
What an agent did, one row per step. A step shimmers while it runs and draws a tick when it finishes; rows with output open to show it, and a group folds the whole run into one line.
How to show a function call in chat — the arguments sent, the result returned, run status, and an expand toggle.
A tool invocation as one quiet row that opens into its JSON, live timer included.
Queued, running, successful, and failed tool activity with a disclosure for input and output.
An agent's tools hung on a board like fishing tackle: slim capability chips at rest, each lifting and paying out a streaming-argument strip the instant its tool is called, with a recency underline that fades over 60 seconds.
An agent's tool call in one line – live timer, status and summary, folds open to highlighted arguments and result with copy, retry on errors and a group that sums them up
One tool call's whole life: input, result, errors and retries.
One call in detail: status and duration, input, output or error as highlighted JSON, retry.
Agent tool call card showing the tool name, argument preview, duration and a running, done or failed status chip, with collapsible monospace output below.
Collapsible record of one tool call: name, status, duration, and the exact input and output on demand.
An AI agent tool call card that shows the arguments, result, duration and status of one call, with a run again action.
A collapsible card for one AI tool call, showing its name, status and duration, with the input and result as formatted JSON.
An agent's tool call in one line – live timer, status and summary, folds open to highlighted arguments and result with copy, retry on errors and a group that sums them up
Compact collapsed tool call row with a rotating chevron and monospace tool name that expands to show its details. Keeps agent traces tidy inside chat threads.
Names a tool call and shows its state and attempt.
Collapsible tool invocation panel with timing and I/O inspection.
Collects search, write, command, and failure states into an inspectable execution record.
An expandable section showing AI agent tool usage with icons, inputs, and outputs.
Debug panel for inspecting AI tool executions with timing, status indicators, and expandable JSON input/output views. Perfect for debugging MCP tools, function calls, and agent actions.
Tool result panel with the tool name, a success, error or running status badge and the output shown in a JSON code block. Built in React for AI agent UIs.
Compact session summary of an agent's tool calls: one row per action on a vertical rail, with the target it touched, its diff stats, and when it ran
Switches for the tools a model may use — web search, code execution, file access — with unmistakable on and off states.
Showing 28,681–28,704 of 29,980
Showing 28,681–28,704 of 29,980